mallam pat i

7/28/2019 Mallam Pat i

http://slidepdf.com/reader/full/mallam-pat-i 1/12

GENERAL ARTICLES

A Systematic Review (Meta-Analysis) of the Accuracy of

the Mallampati Tests to Predict the Difficult Airway

Anna Lee, PhD, MPH, Lawrence T. Y. Fan, MBBS, FANZCA, Tony Gin, MD, FRCA, FANZCA, MB, ChB,Manoj K. Karmakar, MD, FRCA, and Warwick D. Ngan Kee, MD, FANZCA, MB, ChB

Department of Anaesthesia and Intensive Care, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin,NT, Hong Kong

The original and modified Mallampati tests are com-monly used to predict the difficult airway, but there iscontroversy regarding their accuracy. We searchedMEDLINE and other databases for prospective studies

of patients undergoing general anesthesia in which theresults of a preoperative Mallampati test were com-pared with the subsequent rate of difficult airway (dif-ficult laryngoscopy, difficult intubation, or difficultventilation as reference tests). Forty-two studies enroll-ing 34,513 patients wereincluded. The definitions of thereference tests varied widely. For predicting difficultlaryngoscopy, both versions of the Mallampati test had

good accuracy (area under the summary receiver oper-atingcharacteristic (sROC) curve ϭ 0.89 Ϯ 0.05 and0.78Ϯ 0.05, respectively). For predicting difficult intuba-tion, the modified Mallampati test had good accuracy

(area under the sROC curve ϭ 0.83 Ϯ 0.03) whereas theoriginal Mallampati test was poor (area under thesROC curve ϭ 0.58 Ϯ 0.12). The Mallampati tests werepoor at identifying difficult mask ventilation. Publica-tion bias was not detected. Used alone, the Mallampatitests have limited accuracy for predicting the difficultairway and thus are not useful screening tests.

(Anesth Analg 2006;102:1867–78)

D

ifficult laryngoscopy and difficult tracheal intubation occur in 1.5% to 8% of general anesthetics(1). Of available methods, the Mallampati, orig-

inal (2) and modified (3,4) tests, are used as a preop-erative bedside test to predict a difficult airway (5).However, the usefulness of this test is unclear, aspublished studies have produced variable estimates of diagnostic test accuracy. The original Mallampati test(2) identified difficult intubations with a high degreeof accuracy, with sensitivity of 50% and specificity of 100%. However, subsequent larger studies haveshown only modest degrees of accuracy using theoriginal (6) and modified (7–9) versions of the test.Furthermore, the accuracy of the Mallampati test mayvary according to patients’ ethnic group and sex and

whether they are pregnant (10). For example, in Asianpatients it may be more difficult to intubate the tra-chea than in Caucasians (11,12).

The objective of this systematic review was to de-termine the accuracy of the Mallampati test for pre-dicting the difficult airway. For the purposes of thisreview, the definition of a difficult airway includeddifficult laryngoscopy, difficult tracheal intubation,and difficult ventilation. The null hypothesis testedwas that all versions of the Mallampati test had pooraccuracy for identifying difficult airway. We also ex-plored sources of heterogeneity to increase the clinicalrelevance of the results.

MethodsThis systematic review and meta-analysis followed

guidelines on conducting systematic reviews of diag-nostic studies (13,14). We included all prospective ob-servational studies of patients undergoing general an-esthesia who had preoperative Mallampati testassessments and a subsequent assessment of difficultlaryngoscopy, difficult tracheal intubation, or difficultmask ventilation. Difficult airway was defined by agrade III score in the original Mallampati test (2) or agrade III or IV in the modified Mallampati test (3,4)(Table 1).

The studies we assessed included patients with noknown risk factors for difficult tracheal intubation as

Accepted for publication February 6, 2006.Supported solely from departmental and institutional funding.Address correspondence to Anna Lee, PhD, MPH, Department of

Anaesthesia and Intensive Care, The Chinese University of HongKong, Prince of Wales Hospital, Shatin, NT, Hong Kong. Addresse-mail to [email protected].

DOI: 10.1213/01.ane.0000217211.12232.55

©2006 by the International Anesthesia Research Society0003-2999/06 Anesth Analg 2006;102:1867–78 1867



well as patients with upper airway pathology, diabe-tes, obesity, and patients who were pregnant. Alltypes of surgery were considered. Patients undergo-ing indirect laryngoscopy were excluded (8,15). Ret-rospective studies (3,16) and case-control studies (17–

19) were excluded because these would overestimatethe diagnostic test accuracy compared with studiesusing a prospective clinical population. The referencetests for difficult airway with which the Mallampatitest were compared included difficult laryngoscopy(as defined by the four-grade Cormack and Lehanescoring system (20) or the modified five-grade Cormackand Lehane classification (21) and difficult intubationand difficult ventilation (as defined by the authors).

Search Strategy

A systematic search of all relevant prospective ob-servational studies was conducted. Relevant studieswere identified from electronic databases (MEDLINE,EMBASE, Science Citation Index, The Cochrane Li-

brary) January 1985–December 2004, and referencelists of relevant studies and reviews in major journalsrelated to anesthesia. Articles were restricted to thosepublished in English, as there is no evidence to sug-gest a strong association between language restrictionand publication bias in systematic reviews of diagnos-tic tests (22). We used four databases to ensure thatrelevant articles were identified, as publication bias ismore likely to be found if only one to two databases

are used in systematic reviews of diagnostic tests (22).

In addition, the following subject headings and textwords, and their combinations, were included in elec-tronic database search strategy: sensitivity, specificity,screening, false positive, false negative, predictive value of tests, reference values, roc analyses, roc area, roc character-

istics, roc curve, endotracheal intubation, intratracheal in-tubation, laryngoscopy, difficult laryngoscopy, difficult in-tubation, Mallampati and Cormack and Lehane.

The methodological quality of eligible studies wasassessed independently under open conditions. Meth-ods of recruitment and blinding between test andreference test results among anesthesiologists wererecorded. The patient population, type of surgery anddetails of test and reference tests were also collected.Data were obtained from studies independently bytwo or more investigators using a standardized dataextraction form; disagreements were resolved by con-

sensus. The primary author was contacted by letter oremail for relevant data that were not presented in theoriginal publication.

Outcome Measures

The primary outcomes were 1) difficult laryngos-copy (20,21) (Cormack and Lehane Grades 2b, 3 and 4)(Table 1) and 2) difficult tracheal intubation (as there isno standard definition for difficult intubation, we ac-cepted the definition used by authors from eachstudy). The secondary outcome was difficult ventila-tion, as defined by authors from each study. The pri-

mary and secondary outcomes were chosen because

Table 1. Definition of Mallampati Tests and Cormack and Lehane Scales

Original Mallampati Test (2) Grade 1 Faucial pillars, soft palate, and uvula could be visualized

Grade 2 Faucial pillars and soft palate could be visualized, butuvula was masked by base of the tongue

Grade 3 Only soft palate could be visualized

Modified Mallampati Test (3) Class 1 Soft palate, fauces, uvula, pillars seenClass 2 Soft palate, fauces, uvula seenClass 3 Soft palate, base of uvula seenClass 4 Soft palate not visible at all

Modified Mallampati Test (4) Class 0 Ability to see any part of the epiglottis on mouthopening and tongue protrusion

Class 1 Soft palate, fauces, uvula, pillars seenClass 2 Soft palate, fauces, uvula seenClass 3 Soft palate, base of uvula seenClass 4 Soft palate not visible at all

Cormack and Lehane scale (20) Grade 1 Full view of the glottisGrade 2 Partial view of the glottis or arytenoidsGrade 3 Only epiglottis visibleGrade 4 Neither glottis nor epiglottis visible

Modified Cormack and Lehane scale (21) Grade 1 Full view of the glottisGrade 2a Partial view of the glottisGrade 2b Arytenoids or posterior part of the vocal cords only just

visibleGrade 3 Only epiglottis visibleGrade 4 Neither glottis nor epiglottis visible

1868 LEE ET AL. ANESTH ANALGSYSTEMATIC REVIEW OF THE MALLAMPATI TEST 2006;102:1867–78



they are related to consensus guidelines (23) and areclinically important measures of difficult airway.

Statistical Analysis

The sensitivity, specificity, positive likelihood ratio,and negative likelihood ratio were determined indi-vidually from each included study. The accuracy of the test was judged by the magnitude of the positiveand negative likelihood ratios (how much a givendiagnostic test result will increase or decrease thepre-test probability of the target disorder) using theguide by Jaeschke et al. (24). The potential problemsassociated with sensitivities and specificities of 0%and 100% were solved by adding 0.5 to all cells of thediagnostic 2 ϫ 2 table (13). The DerSimonian Lairdmethod (random effects model) was used to incorpo-rate variation among studies when pooling sensitivity,

specificity, positive likelihood ratio, and negative like-lihood ratio. However, when there was an association

between sensitivity and specificity across studies(threshold effect), we did not report the individualweighted average for sensitivity, specificity, positivelikelihood ratio, and negative likelihood ratio, as thiswould lead to under-estimation of diagnostic test per-formance (25,26). Instead, a summary receiver opera-tor characteristic (sROC) curve of all the studies wascreated (27), as this is a better summary of the studyresults than a single joint summary estimate of sensi-tivity and specificity (see Appendix for details aboutconstruction and interpretation of sROC curve). We

used the area under the sROC curve to judge thedegree of accuracy of the tests according to publishedguidelines (28) (Ն0.97 ϭ excellent, 0.93 to 0.96 ϭ verygood, 0.75 to 0.92 ϭ good, 0.50 to 0.75 ϭ poor). Wecomputed a weighted average of the specificity fromall studies using the random effects models; then thesensitivity was calculated from the sROC curve equa-tions. Positive and negative likelihood ratios were de-rived from the summary sensitivity and specificity.

Heterogeneity was described using the I 2 statistic(29) for pooling sensitivity, specificity, and positiveand negative likelihood ratios. Sensitivity analyseswere performed to evaluate the robustness of resultsaccording to blinding of test results among anesthesi-ologists (blinding versus unclear/no blinding) for theprimary outcomes.

Publication bias was assessed using the Egger’sweighted regression method (30) with precision (1/standard error) and log odds ratio plotted. The inter-cept value in Egger’s regression method provides anestimate of asymmetry of the funnel plot, with posi-tive values indicating a trend towards higher levels of test accuracy in studies with smaller sample sizes. Thethreshold of significance was set at P Ͻ 0.10 for thismethod as this test has low power (31). All statistical

analyses were performed using Stata version 8.0 (Stata

Corp, College Station, TX) and MetaDiSc version 1.1.1(Zamora J, Muriel A, Abraira V, Madrid, Spain).

Results

Our literature search identified 42 studies that en-rolled 34,513 patients (2,4,6–9,11,12,32–65). One studywas excluded because of inconsistencies in the pre-sented data (58). Two studies were excluded in whichthe modified Mallampati test was assessed as part of amore comprehensive risk score (53,57) but data for themodified Mallampati test component were unavail-able. The characteristics of the included studies aresummarized in Tables 2 and 3. There were no studiesin children.

The quality of the studies was assessed according tothe method of patient recruitment and blinding of theMallampati tests and reference tests results among anes-

thesiologists. Patients were recruited consecutively in 19studies (2,4,6,8,34,35,40– 44,46,47,49,50,52,59,60,62). Con-stantikes (39) took a convenient sample of 30 pa-tients. Blinding was reported in 10 studies(7,11,35,38,46,48–50,52,53). Both consecutive patientrecruitment and blinding occurred in 5 studies(35,46,49,50,52).

There was poor documentation about how the Mal-lampati tests were done with regard to body and headpositions and the use of phonation. Nine studies hadadequate description of all three aspects(4,7,43,51,53,56,57,64,65). Phonation during the Mal-

lampati test was described in six studies(33,39,42,44,53,65). The modified Mallampati test wasperformed in Asian patients in eight studies(8,11,12,46,55,56,59,64). The modified Mallampati testwas used in obstetric patients in five studies(7,11,43,46,59). In two studies (11,59) separate data of test characteristics were given for obstetric and gyne-cological patients. As there was a discrepancy be-tween the abstract and text in the proportion of ob-stetric and non-obstetric patients in one study (52), itwas excluded from meta-regression analyses.

Difficult Laryngoscopy

All studies used Cormack and Lehane’s originalclassification for defining difficult laryngoscopy ex-cept one study (55) that used the modified five-gradescore (21).

The original Mallampati test was used in 9 studieswith 14,438 patients (Table 2). The prevalence of dif-ficult laryngoscopy ranged from 6% to 27%. There wasa high prevalence of difficult laryngoscopy in patientswith cervical disease (36) and in patients with diabetes(40). In one study, the authors attributed the highprevalence of difficult laryngoscopy to the use of theMcCoy laryngoscope (37). The sensitivity and speci-

ficity of the individual studies ranged from 0.05 to 1.00

ANESTH ANALG LEE ET AL. 18692006;102:1867–78 SYSTEMATIC REVIEW OF THE MALLAMPATI TEST



and 0.65 to 0.98, respectively (Table 2). The positiveand negative likelihood ratios of the 9 studies rangedfrom 1.71 to 32.08 and 0.14 to 0.97, respectively (Table2). As there was an apparent relationship betweensensitivity and specificity (Spearman r ϭ 0.45), a sROCcurve was constructed (Fig. 1). The area (Ϯ se) underthe symmetrical sROC curve was 0.89 (0.05). Consid-

ering the threshold effect, the diagnostic odds ratio(DOR) was 19.57 (95% CI, 5.02 to 76.27). The summaryestimate for sensitivity was 0.71 derived from equa-tion (2) for the sROC curve at the summary specificityof 0.89. Hence, the summary positive and negativelikelihood ratios were 6.45 and 0.33, respectively.Blinding (relative DOR) (rDOR) 0.49; 95% confidenceinterval [CI], 0.02 to 12.72) and phonation (rDOR 0.11,95% CI, 0.00 to 5.47) did not change the diagnosticperformance of the original Mallampati test for pre-dicting difficult laryngoscopy. There was no evidenceof publication bias being present in this meta-analysis

(tϭ

1.19, Pϭ

0.27).

The modified Mallampati tests (3,4) were used in 19studies with 10,579 patients (Table 3). There was widevariability in the prevalence of difficult laryngoscopy,which ranged from 2% to 26% (Table 2). The highestprevalence was in acromegaly patients (51). The sen-sitivity and specificity of the individual studies rangedfrom 0.12 to 1.00 and 0.44 to 0.98, respectively (Table

2). There was an association between sensitivity andspecificity (Spearman’s r ϭ 0.32). The area (Ϯ se)under the symmetrical sROC curve (Fig. 2) was 0.78(0.05). Considering the threshold effect, the DOR was6.45 (95% CI, 2.73 to 15.22). The summary estimate forsensitivity was 0.55 and the summary specificity was0.84. The summary positive and negative likelihoodratios were 3.44 and 0.54, respectively. There was nosignificant difference in the areas under the sROCcurve between the original and modified versions of the Mallampati test (z ϭ 1.56; P ϭ 0.12). Publication

bias was not evident in the meta-analysis of 19 studies

for difficult laryngoscopy (tϭ Ϫ

0.57; Pϭ

0.57).

Table 2. Prospective Studies of Original Mallampati Test with Three Reference Tests for Difficult Airway

StudyPatients

(n)Women,

(%)

Age

mean(range)

(yr)

Prevalenceof

difficultairway

accordingto

referencestandard

(%)

Truepositive

(n)

Falsepositive

(n)

Falsenegative

(n)

TrueNegative

(n)Sensi-tivity

Speci-ficity

PositiveLikihood

Ratio

NegativeLikelihood

Ratio

Difficult laryngoscopyBilgin et al.,

1998 (32)500 51 46 (16-80) 9 39 12 7 442 0.85 0.97 32.08 0.16

Calder et al.,1995 (36)

251 51 54 (8-84) 20 29 8 22 192 0.57 0.96 14.22 0.45

Cohen et al.,1992 (33)

663 NS NS (18-88) 6 21 55 29 558 0.42 0.91 4.68 0.64

Constantikes,1993 (39)

30 NS NS (18-NS) 7 2 8 0 20 1.00 0.71 2.84 0.24

El-Ganzouri etal., 1996 (6)

10507 NS NS (18-NS) 6 287 1080 355 8785 0.45 0.89 4.08 0.62

Nadal et al.,1998 (40)

83 64 53 (18-NS) 27 1 1 21 60 0.05 0.98 2.77 0.97

Randell et al.,1998 (37) 100 38 48 (NS) 20 3 7 17 73 0.15 0.91 1.71 0.93

Tse et al., 1995(35)

471 53 NS (19-89) 13 41 145 21 264 0.66 0.65 1.87 0.53

Voyagis et al.,1998 (34)

1833 NS NS 8 134 227 18 1454 0.88 0.87 6.53 0.14

Difficult intubationBilgin et al.,

1998 (32)500 51 46 (16-80) 8 17 34 23 426 0.43 0.93 5.75 0.62

Cohen et al.,1992 (33)

663 NS NS (18-88) 11 25 51 49 538 0.34 0.91 3.90 0.73

El-Ganzouri etal., 1996 (6)

10507 NS NS (18-NS) 6 287 1080 355 8785 0.45 0.89 4.08 0.62

Mallampati etal., 1985 (2)

210 78 39 (18-82) 13 14 1 14 181 0.50 1.00 91.0 0.50

Tse et al., 1995(35)

471 53 NS (19-89) 13 41 145 21 264 0.66 0.65 1.87 0.53

Difficult ventilationEl-Ganzouri et

al., 1996 (6)10507 NS NS (18-NS) 0.1 3 1364 5 9135 0.38 0.87 2.89 0.72

NS ϭ not specified.




Blinding (rDOR 1.49; 95% CI, 0.22 to 10.31) andphonation (rDOR 0.86; 95% CI, 0.13 to 5.76) did notchange the diagnostic performance of the modified

Mallampati test for predicting difficult laryngoscopy.

Also, there were no differences in diagnostic test per-formance between studies of Asian and Caucasianpatients (rDOR 1.09; 95% CI, 0.21 to 5.64). However,

on meta-regression, the modified Mallampati test was

Table 3. Prospective Studies of Modified Mallampati Test with Three Reference Tests for Difficult Airway

Study, Year(Reference)

Patients(n)

Women,(%)

Age mean(range)

(yr)

Prevalenceof difficult

airwayaccording to

referencestandard (%)

Trueposi-tive(n)

Falseposi-tive(n)

Falsenega-tive(n)

Truenega-tive(n)

Sensi-tivity

Speci-ficity

PositiveLiki-hoodRatio

NegativeLikeli-hoodRatio

Difficult laryngoscopyAdnet et al., 2001 (41) 1171 57 NS (15-90) 9 25 53 80 1013 0.24 0.95 4.79 0.80Bouaggard et al., 2004

(42)320 88 NS (18-NS) 10 9 6 23 282 0.28 0.98 13.50 0.73

Brodsky et al., 2002(44)

100 78 44 (NS) 9 5 28 4 63 0.56 0.69 1.81 0.64

Butler et al., 1992 (12) 220 61 NS (16-80) 8 10 38 8 164 0.56 0.81 2.95 0.55Cattano et al., 2004

(9)1956 NS NS (18-NS) 2 15 182 28 1731 0.35 0.91 3.67 0.72

Constantikes, 1993(39)

30 NS NS (18-NS) 7 2 15 0 13 1.00 0.46 1.56 0.36

Ezri et al., 2001 (4) 764 48 44 (18-NS) 11 68 196 13 487 0.84 0.71 2.93 0.23Ezri et al., 2003 (60) 1472 49 NS (18-NS) 10 116 354 36 966 0.76 0.73 2.85 0.32Ezri et al., 2003 (61) 50 58 NS (NS) 18 3 6 6 35 0.33 0.85 2.28 0.78Ezri et al., 2003 (62) 644 43 NS (40-NS) 7 11 208 32 393 0.26 0.65 0.74 1.14

Gercek et al., 2003(38) 500 63 NS (NS) 5 3 14 22 461 0.12 0.97 4.07 0.91

Gupta et al., 2003 (46) 372 100 25 (16-37) 7 15 8 10 339 0.60 0.98 26.03 0.41Ita et al., 1994 (45) 57 68 31 (NS) 5 2 2 1 52 0.67 0.96 18.00 0.35Koh et al., 2002 (55) 605 56 NS (16-NS) 5 14 45 17 529 0.45 0.92 5.76 0.60Merah et al., 2004 (43) 80 100 31 (NS) 10 7 3 1 69 0.88 0.96 21.00 0.13Rocke et al., 1992 (7) 1500 100 26 (NS) 2 16 381 11 1092 0.59 0.74 2.29 0.55Schmitt et al., 2000

(51)128 53 46 (19-78) 26 25 53 8 42 0.76 0.44 1.36 0.55

Vani et al., 2000 (64) 50 56 57 (18-NS) 16 1 4 7 38 0.13 0.91 1.31 0.97Yeo et al., 1992 (59) 560 100 NS (15-69) 2 3 25 8 524 0.27 0.95 5.99 0.76Difficult intubationAdnet et al., 2001 (41) 1171 57 NS (15-90) 8 24 54 66 1027 0.27 0.95 5.34 0.77Arne et al., 1998 (47) 1200 NS NS (15-NS) 4 39 168 11 982 0.78 0.85 5.34 0.26Ayuso et al., 2003 (50) 181 26 54 (NS) 30 29 26 25 101 0.54 0.80 2.62 0.58Bouaggard et al., 2004

(42)

320 88 NS (18-NS) 5 7 8 10 295 0.41 0.97 15.60 0.60

Brodsky et al., 2002(44)

100 78 44 (NS) 12 7 26 5 62 0.58 0.71 1.97 0.59

Cattano et al., 2004 (9) 1956 NS NS (18-NS) 3 28 169 28 1731 0.50 0.91 5.62 0.55Frerk et al., 1991 (54) 244 41 44 (18-82) 5 9 43 2 190 0.82 0.82 4.43 0.22Gercek et al., 2003

(38)500 63 NS (NS) 7 2 15 31 452 0.06 0.97 1.89 0.97

Gupta et al., 2003 (46) 372 100 25 (16-37) 7 15 8 9 340 0.63 0.98 27.19 0.38Iohom et al., 2003 (48) 212 49 NS (18-NS) 9 8 22 12 170 0.40 0.89 3.49 0.68Ita et al., 1994 (45) 57 68 31 (NS) 2 0 4 1 52 0 0.93 3.17 0.81

Juvin et al., 2003 (63) 263 72 NS (18-79) 9 20 75 3 165 0.87 0.69 2.78 0.19Kaul et al., (56) 500 42 36 (18-87) 8 31 24 11 434 0.74 0.95 14.09 0.28Khan et al., 2003 (49) 300 NS NS (16-NS) 6 14 94 3 189 0.82 0.67 2.48 0.26Merah et al., 2004 (43) 80 100 31 (NS) 10 7 3 1 69 0.88 0.96 21.00 0.13Rocke et al., 1992 (7) 1500 100 26 (NS) 2 19 378 13 1090 0.59 0.74 2.31 0.55Savva, 1994 (52) 350 57 NS (18-81) 5 11 113 6 220 0.65 0.66 1.91 0.53

Wong et al., 1999 (11) 411 100 NS (NS) 2 6 150 1 254 0.86 0.63 2.29 0.23Yamamoto et al., 1997

(8)3680 49 52 (NS) 2 38 1723 18 1901 0.68 0.53 1.43 0.61

Yeo et al., 1992 (59) 560 100 NS (15-69) 2 3 25 8 524 0.27 0.95 5.99 0.76Difficult ventilationCattano et al., 2004

(9)1956 NS NS (18-NS) 2 15 179 32 1703 0.32 0.91 3.36 0.75

Langeron et al., 2000(65)

1502 55 NS (18-NS) 5 17 186 58 1241 0.23 0.87 1.74 0.89

NS ϭ not specified.




5.08 (95% CI, 1.26 to 20.58; P ϭ 0.03) times moreaccurate in studies of obstetric patients than in studiesin surgical patients.

Difficult Tracheal IntubationThere was wide variation in the definition of difficult

tracheal intubation. Many studies (6,8,11,35,43,49,56,59)used the original Cormack and Lehane definition (20).Three studies (41,42,63) defined difficult tracheal intu-

bation as a score Ͼ5 on the Intubation Difficulty Scoredescribed by Adnet et al. (41). This scoring systemincorporates the number of attempts, number of ad-ditional operators, number of alternative intubationtechniques, Cormack and Lehane grade, liftingforce, laryngeal pressure, and vocal cord mobility.Two studies (7,46) in obstetric patients used Rockeet al.’s classification (7) for defining difficult tra-cheal intubation.

The original Mallampati test (2) was used to predictdifficult tracheal intubation in 5 studies enrolling12,351 patients (Table 2). The prevalence of difficulttracheal intubation ranged from 6% to 13%. Therewere low sensitivities (0.34 to 0.66) and varying spec-ificities (0.65 to 1.00). The positive likelihood ratiosvaried from 1.87 to 91.0; negative likelihood ratiosvaried from 0.50 to 0.73. There was an association

between sensitivity and specificity (Spearman’s r ϭ

0.30). As the DOR was not constant across the thresh-old (b ϭ Ϫ0.71; 95% CI, Ϫ1.21 to Ϫ0.21), the sROC

curve was asymmetrical (Fig. 3). The area (Ϯ se

) under

the asymmetrical sROC was 0.58 (0.12). The summaryestimate for the sensitivity was 0.50 derived from theequation (3) for the sROC curve at the summary spec-ificity of 0.89. Phonation during the test (rDOR 0.27;95% CI, 0.00 to 15.55) and blinding (rDOR 15.65; 95%

CI, 0.14 to 1784.76) did not appear to affect the overallaccuracy of the test. There was no evidence of publi-cation bias being present in the studies pooled (t ϭ

Ϫ0.39, P ϭ 0.72).Twenty studies enrolling 13,957 patients (Table 3)

examined the use of the modified Mallampati (3,4)test for predicting difficult tracheal intubation. Theprevalence of difficult tracheal intubation rangedfrom 2% to 30% (Table 3). The high prevalence of difficult tracheal intubation (30%) occurred in pa-tients with pharyngolaryngeal disease (50). The sen-sitivities ranged from 0 to 0.88 and the specificitiesranged from 0.53 to 0.98 (Table 3). The positivelikelihood ratios varied 1.43 to 27.19; negative like-lihood ratios varied from 0.13 to 0.97 (Table 3).There was a correlation between sensitivity andspecificity (Spearman’s r ϭ 0.47). The area (Ϯ se)under the symmetrical sROC (Fig. 4) was 0.83 (0.03).Considering the threshold effect, the DOR was 10.43(95% CI, 5.32 to 20.48). The summary estimate forsensitivity was 0.76 when the summary specificitywas 0.77. Hence, the summary positive and negativelikelihood ratios were 3.30 and 0.31, respectively.The modified Mallampati test was better at identi-fying difficult tracheal intubation than the original

Mallampati test (zϭ

2.02; Pϭ

0.04). There was

Figure 2. Summary receiver operating characteristic curve of themodified Mallampati test for predicting difficult laryngoscopy. Thesummary estimates of sensitivity and specificity are 0.55 and 0.84,respectively.

Figure 1. Summary receiver operating characteristic curve of theoriginal Mallampati test for predicting difficult laryngoscopy. Eachcircle represents the results of a single study. The summary esti-mates of sensitivity and specificity are 0.71 and 0.89, respectively.




no evidence of publication bias in the 20 studiespooled for difficult tracheal intubation (t ϭ 0.89; P ϭ

0.39).Blinding (rDOR 0.52; 95% CI, 0.17 to 1.65), phona-

tion (rDOR 0.94; 95% CI, 0.13 to 6.69), and studies of Asian patients (rDOR 1.65; 95% CI; 0.44 to 6.24) didnot change the diagnostic performance of the modi-fied Mallampati test for predicting difficult trachealintubation. The difference between obstetric patientsand non-obstetric patients in the accuracy of the mod-ified test for predicting difficult tracheal intubationwas not significant (rDOR 2.69; 95% CI, 0.81 to 8.88,P ϭ 0.10).

No failed intubation occurred in 12 studies(4,12,31,34,39– 41,46– 48,50,62). When failed intuba-tion occurred, the prevalence varied from 0.1% (7) to3.8% (57).

Difficult Ventilation

Three studies recorded difficult ventilation (6,9,65).Definitions varied and included inability to obtainchest excursion sufficient to maintain a clinically ac-ceptable capnogram waveform despite optimal headand neck positioning, use of muscle paralysis, use of an oral airway, and optimal application of a facemask(6). In another study (65), ventilation via a mask wasconsidered difficult only when the anesthesiologistconsidered that the difficulty was clinically relevant

and could have led to problems if mask ventilation

had to be maintained for a longer time. Bag-maskventilation was considered difficult if one or more of the following factors were present: inability to main-tain an adequate seal; inability to obtain chest excur-sion, obtain a good capnograph tracing, or maintain

oxygen saturation more than 90% despite good musclerelaxation; the necessity of using a Guedel oral airway;or two-person bag-mask ventilation (9). The sensitiv-ity, specificity, and positive and negative likelihoodratios are shown in Tables 2 and 3 for this outcome.Pooled sensitivity and specificity of the modified Mal-lampati test (3) were 0.26 (95% CI, 0.19 to 0.35) and0.89 (95% CI, 0.88 to 0.90), respectively (9,65). Therewas little heterogeneity between the two studies forsensitivity (I 2 ϭ 21%). However, there was substantialheterogeneity for specificity (I 2 ϭ 90%). The positiveand negative likelihood ratios were 2.42 (95% CI, 1.25to 4.66) and 0.83 (95% CI, 0.71 to 0.98), respectively,suggesting poor accuracy. There were moderateamounts of heterogeneity among studies for positiveand negative likelihood ratios (I 2 ϭ 78% and 52%respectively).

To put the results of this systematic review in aclinical context, readers can estimate the post-testprobability of a difficult airway after an examinationof the airway using the modified Mallampati test ac-cording to the prevalence of difficult airway in theirpopulation (Fig. 5). The range of pre-test probabilitiesreflects the range of prevalence reported in this sys-tematic review. If the pre-test probability of difficult

airway is 10%, a positive test generates a post-test

Figure 3. Summary receiver operating characteristic curve of theoriginal Mallampati test for predicting difficult tracheal intubation.Each circle represents the results of a single study. The summaryestimates of sensitivity and specificity are 0.50 and 0.89,respectively.

Figure 4. Summary receiver operating characteristic curve of themodified Mallampati test for predicting difficult tracheal intubation.The summary estimates of sensitivity and specificity are 0.76 and0.77, respectively.




probability of difficult laryngoscopy of 28% and diffi-cult intubation of 27%; a negative test generates apost-test probability of difficult laryngoscopy of 6%and difficult intubation of 3%.

DiscussionThis systematic review of the literature identifiedmany studies describing the performance of the orig-inal (2) and modified (3,4) Mallampati tests to predictthe difficult airway. There was substantial variabilityin the reported sensitivity and specificity among thestudies and in definitions of the reference tests. Unlikemeta-analysis of interventions, which produces oneanswer, the performance of diagnostic tests is affected

by changes in sensitivity and specificity as reflected inthe sROC curves (Figs. 1 to 4). Thus, there is no unique

joint summary estimate of sensitivity and specificity; itis only possible to obtain a summary estimate of onevalue conditional on the value of the other (66). Over-all, the accuracy of the Mallampati tests was poor togood, depending on the version of the test and refer-ence test used. Our results are not directly comparableto a recent meta-analysis of bedside screening test forpredicting difficult intubation (67). In Shiga et al.’smeta-analysis (67), there was no distinction made be-tween the various versions of the Mallampati test or

between difficult laryngoscopy and difficult intuba-tion, a major limitation of their study. Nevertheless,

they concluded that the Mallampati test’s clinical

value of a bedside screening test was limited as it hadpoor to moderate discriminative power when usedalone. Our results concur with this view.

Both versions of the Mallampati test had good ac-curacy for identifying difficult laryngoscopy as as-

sessed according to the original and modified Cor-mack and Lehane grading system. This system iswidely used in clinical practice to describe the bestview obtained by direct laryngoscopy with or withoutmanipulation of the larynx. However, there is consid-erable uncertainty and inaccuracy in this grading sys-tem, especially between grade 2 and grade 3 (68). Theincidence of difficult laryngoscopy may be underesti-mated, as most of the studies used the original Cor-mack and Lehane grading system. Approximately 3%(55) to 7% (21) of patients graded 2b, who wouldotherwise have been rated grade 2 in the originalsystem, will have a high risk of difficult laryngoscopy.

Such misclassification may affect the overall test per-formance of the Mallampati tests. Many studies usedthe same Cormack and Lehane grading system todefine both difficult laryngoscopy and difficult intu-

bation. Although difficult intubation is the end resultof difficult laryngoscopy, the former also depends onthe operator’s experience, patient characteristics, andclinical setting.

The recommended best way to perform the Mallam-pati test for predicting difficult laryngoscopy is put-ting the patient in a sitting position, with the head infull extension, the tongue out, and with phonation

(53). However, many studies did not specifically doc-ument the way the Mallampati test was performed.Therefore, variations in the conduct of Mallampatitests may contribute to some of the heterogeneity of results seen in this systematic review. Unexpectedly,phonation did not influence the overall accuracy of theMallampati tests.

There was a large variation among studies in thedefinition of difficult tracheal intubation. There is nocurrent consensus on the definition of difficult tra-cheal intubation. Therefore, we used the definitionfrom each study to establish an operational referencestandard reflecting current clinical practice. The dif-

ferent definitions of difficult tracheal intubation mayexplain, in part, the heterogeneity of results in thesROC curves. For predicting difficult tracheal intuba-tion, the original Mallampati test had very poor accu-racy. Four of the five studies had sensitivities Ͻ50%.Small increases in sensitivity led to large sacrifices inspecificity. The asymmetrical sROC curve suggeststhat accuracy was dependent on threshold. The lowestaccuracy occurred when the threshold was high. Thismay be related to the quality of study. The lowestaccuracy occurred in a study with the least amount of reviewer and patient selection bias (35). In contrast,

the modified Mallampati test had good accuracy for

Figure 5. Post-test probability of difficult laryngoscopy and difficulttracheal intubation after using the modified Mallampati test. Post-test probabilities are shown as a function of pretest probability forpositive (Grade III or IV) and negative results (Grade 0, I or II) onthe modified Mallampati test.




predicting difficult tracheal intubation and was signif-icantly better than the original test. The discrepancy inresults between the two versions of the Mallampatitest may be related to the definition of difficult tra-cheal intubation used and difference in the study

populations.The accuracy of the modified Mallampati test forpredicting difficult laryngoscopy was five timeshigher in obstetric patients than in non-obstetricpatients, although for predicting difficult trachealintubation, the difference was not significant. This isconsistent with studies that showed that pregnancycaused a 34% increase in Mallampati grade 4 (10)and that the risk of difficult intubation in obstetricpatients was approximately 8 times more than insurgical patients (3). More difficult laryngoscopy inobstetric patients most likely occurs because of fa-cial and pharyngeal edema secondary to hormon-

ally induced fluid retention (69). These results sug-gest that the Mallampati tests are probably better atpredicting difficult laryngoscopy associated withsoft tissue changes compared with other anatomicalfactors.

We found little evidence of ethnic differences in theaccuracy of modified Mallampati tests for difficultlaryngoscopy and difficult intubation, despite knowncephalometric differences among ethnic populations(70).

In a recent editorial, Murphy et al. (71) suggestedthat we should focus on “ventilatability” rather than“intubatability.” The accuracy of the Mallampati tests

for predicting difficult mask ventilation was poor, butthis was based on only three studies. Therefore, theseresults should be interpreted with caution. For pre-dicting difficult mask ventilation, the presence of 2 of 5 factors (age older than 55 years, body mass indexϾ26 kg/m2, lack of teeth, presence of beard, and his-tory of snoring) was associated with good accuracy(area under the curve 0.76 Ϯ 0.11) (65). As expected,there was a strong association between difficult tra-cheal intubation and difficult mask ventilation (65).

Systematic review and meta-analysis are consideredto provide the least biased estimates of effect but if the“raw material” is flawed, then the conclusions of sys-tematic reviews will be compromised and invalid (66).The quality of reporting varied among studies; only afew studies described study methodology and Mal-lampati test assessments in adequate detail. We as-sumed that the quality of the study was inadequate if it was clearly stated that there were deficiencies indesign and conduct. Omission of reporting specificdetails of a study was associated with systematic dif-ferences in results (72). Of the 42 studies included inthis systematic review, only 5 studies recruited pa-tients consecutively with test results blinded amonganesthesiologists. This suggests that the majority of

studies included in this systematic review may have

less than adequate study methodology. Future studiesof tests for identifying difficult airway should adoptthe Standards for Reporting of Diagnostic Accuracyguidelines (73). This would allow readers to assess thepotential for bias in the study and to evaluate the

generalizability of study results.Interpreting the reference test with knowledge of the results of the test under study can lead to anover-estimation of a test’s accuracy (72). This is knownas review bias. Unblinded studies tend to overesti-mate the diagnostic test accuracy by 1.3 times (95% CI,1.0 to 1.9) (72). However, we did not find a significanteffect of blinding on the Mallampati tests’ accuracy.We also minimized spectrum bias (study sample doesnot include the complete spectrum of patient charac-teristics) by excluding case-control studies from thesystematic review. Diagnostic accuracy can be overes-timated by 3 times (95% CI, 2.0 to 4.5) if the test is

evaluated in a group of patients already known tohave the disease and a separate group of normal pa-tients, as in case-control studies (72).

Publication bias in meta-analyses of test accuracy ishighly prevalent (22). This type of bias is a threat tothe validity of meta-analysis as it can lead to inappro-priate decision making and health care policies. Weundertook a comprehensive literature search usingseveral electronic databases. Although we restrictedour systematic review to include English languagestudies, the inclusion of non-English language studieswould only increase the precision without affecting

the overall accuracy estimates. A previous studyshowed no relationship between publication bias andlanguage restriction in reviews (22). We believe thatour results are robust, as publication bias was notpresent.

As there were no pediatric studies, the results of our systematic review are applicable only to adults.There was a wide range of difficult airway preva-lence, reflecting various patient characteristics, in-cluding pregnancy (7,11,43,46,59), pharyngolaryn-geal disease (50), acromegaly (51), and obesity(44,61). As post-test probability depends on the dis-ease prevalence, knowledge of the prevalence of

difficult laryngoscopy and difficult intubation atany individual hospital will aid in the application of our results (Fig. 5). The decision to perform addi-tional radiographic evaluation, consultationwith other specialists or use special techniques/equipment to manage difficult airways will dependon how high the post-test probability is and, conse-quently, at what level the treatment threshold is set

by the individual anesthesiologist.The results of our systematic review question the

routine use of the Mallampati tests. Given the poorto moderate inter-observer reliability of the modi-

fied Mallampati test (74,75) and the poor to good




accuracy of the Mallampati tests, should we aban-don their use? To decide this, anesthesiologistsshould balance the cost of failing to predict a diffi-cult airway when there is a false negative resultversus the possibility of unnecessary treatment

when there is a false positive result. Used alone, theMallampati tests are insufficient to confidently pre-dict the presence or absence of a difficult airway; we

believe they should form only a limited part of theoverall assessment of the airway. As recommended

by the American Society of Anesthesiologists TaskForce on the management of the difficult airway(23), dentition, thyromental distance, and neck ex-tension are other parts of the airway examinationthat also need to be examined.

Appendix The sROC curve method considers heterogeneityacross studies attributable to differences in the thresh-old values used. Even if the same threshold has beenused, inter-observer differences in the Mallampatigrades (74,75) may lead to inherent variations in thepositive results cutoff. To confirm that there was athreshold effect, the true-positive rate (TPR) and false-positive rate (FPR) of each study were plotted againsteach other, and the Spearman correlation coefficientwas calculated. In creating the sROC, the TPR and FPRwere converted to their logits, and the sum and dif-ferences of the logits were estimated. Equally un-

weighted least squares linear regression of the follow-ing model was performed:

D ϭ a ϩ bS (1)

where D ϭ logit TPR – logit FPR, S ϭ logit TPR ϩ logitFPR, a ϭ intercept term, and b ϭ regression coefficientfor S. D is equivalent to the diagnostic odds ratio(DOR), which conveys the test’s accuracy in discrim-inating diseased subjects from nondiseased subjects(76). S can be interpreted as a measure of the diagnos-tic test threshold, with high values corresponding toliberal inclusion criteria for diseased subjects (76). Theregression coefficient b represents the dependence of the test accuracy on threshold. If b Ϸ 0, then thestudies are homogeneous and can be summarized byan overall DOR noting that a ϭ ln(DOR) (76), giving asymmetrical sROC. The studies are heterogeneouswith respect to the diagnostic odds ratio if b 0 (76).In this case, the sROC is asymmetrical. The DOR isrelated to the area under the sROC curve. A DOR of 1is equivalent to an area under the sROC curve of 50%;the larger the DOR, the larger the area under the sROCcurve. Ninety-five percent confidence intervals (95%CI) were estimated around the DOR. The areas underthe sROC curve of the original and modified Mallam-

pati test were compared using the method outlined by

Hasselblad and Hedges (77). The resulting equations below (equations 2 and 3) represent the logit form of the sROC curve, from which a pooled estimate of TPRand FPR can be obtained. The equation (66) of thecorresponding symmetrical sROC curve is given by:

sensitivity ϭ1

1 ϩ1

DORxͩ1 Ϫ specificity

specificityͪ

(2)

The equation (26) of the corresponding asymmetri-cal sROC curve is given by:

sensitivity ϭ1

1 ϩ1

ea/(1Ϫ

b)xͩ1 Ϫ specificity

specificity ͪ͑ 1ϩ b͒ /͑ 1Ϫ b͒

(3)

Meta-regression was used to explore possible reasonsfor heterogeneity with a priori subgroups, includingtype of patient population (coded 1 ϭ Asians, 0 ϭ

Caucasians; 1 ϭ obstetrics, 0 ϭ non-obstetrics) andphonation (coded 1 ϭ yes, 0 ϭ no) during the Mal-lampati tests. This was done by extending the sROCmodel introduced above (equation 1) to include acovariate (27). The resulting parameter estimates of the covariate can be interpreted, after antilogarithm

transformation, as the relative DOR (rDOR) (72) andreflects the differences in threshold choice at differentlevels of the covariate. Fitting a covariate to the modeldoes not result in a separate sROC curve for each levelof the covariate, as the relationship between TPR andFPR is reflected only in a (78).

The authors thank the authors of the original studies who re-sponded to our requests for unpublished and additional data.

References1. Crosby ET, Cooper RM, Douglas MJ, et al. The unanticipated

difficult airway with recommendations for management. Can JAnaesth 1998;45:757–76.

2. Mallampati SR, Gatt SP, Gugino LD, et al. A clinical sign topredict difficult tracheal intubation: a prospective study. CanAnaesth Soc J 1985;32:429–34.

3. Samsoon GL, Young JR. Difficult tracheal intubation: a retro-spective study. Anaesthesia 1987;42:487–90.

4. Ezri T, Warters RD, Szmuk P, et al. The incidence of class “zero”airway and the impact of Mallampati score, age, sex, and bodymass index on prediction of laryngoscopy grade. Anesth Analg2001;93:1073–5.

5. Mellado PF, Thunedborg LP, Swiatek F, Kristensen MS. Anaes-thesiological airway management in Denmark: assessment,equipment and documentation. Acta Anaesthesiol Scand 2004;

48:350–4.




6. El-Ganzouri AR, McCarthy RJ, Tuman KJ, et al. Preoperativeairway assessment: predictive value of a multivariate risk index.Anesth Analg 1996;82:1197–204.

7. Rocke DA, Murray WB, Rout CC, Gouws E. Relative risk anal-ysis of factors associated with difficult intubation in obstetricanesthesia. Anesthesiology 1992;77:67–73.

8. Yamamoto K, Tsubokawa T, Shibata K, et al. Predicting difficult

intubation with indirect laryngoscopy. Anesthesiology 1997;86:316–21.

9. Cattano D, Panicucci E, Paolicchi A, et al. Risk factors assess-ment of the difficult airway: an Italian survey of 1956 patients.Anesth Analg 2004;99:1774–9.

10. Pilkington S, Carli F, Dakin MJ, et al. Increase in Mallampatiscore during pregnancy. Br J Anaesth 1995;74:638–42.

11. Wong SH, Hung CT. Prevalence and prediction of difficultintubation in Chinese women. Anaesth Intensive Care 1999;27:49–52.

12. Butler PJ, Dhara SS. Prediction of difficult laryngoscopy: anassessment of the thyromental distance and Mallampati predic-tive tests. Anaesth Intensive Care 1992;20:139–42.

13. Deville WL, Buntinx F, Bouter LM, et al. Conducting systematicreviews of diagnostic studies: didactic guidelines. BMC MedRes Methodol 2002;2:9.

14. Irwig L, Tosteson AN, Gatsonis C, et al. Guidelines for meta-analyses evaluating diagnostic tests. Ann Intern Med 1994;120:667–76.

15. Sonbul ZM, Demian AD, Deiab DG, El Asfour AA. Prediction of difficult intubation in obstructive sleep apnoea patients. Egyp-tian J Anaesth 2004;20:189–94.

16. Naguib M, Malabarey T, AlSatli RA, et al. Predictive models fordifficult laryngoscopy and intubation: a clinical, radiologic andthree-dimensional computer imaging study. Can J Anaesth1999;46:748–59.

17. Frerk CM, Till CB, Bradley AJ. Difficult intubation: thyromentaldistance and the atlanto-occipital gap. Anaesthesia 1996;51:738–40.

18. Samra SK, Schork MA, Guinto FC Jr. A study of radiologicimaging techniques and airway grading to predict a difficultendotracheal intubation. J Clin Anesth 1995;7:373–9.

19. Karkouti K, Rose DK, Wigglesworth D, Cohen MM. Predictingdifficult intubation: a multivariable analysis. Can J Anaesth2000;47:730–9.

20. Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics.Anaesthesia 1984;39:1105–11.

21. Yentis SM, Lee DJ. Evaluation of an improved scoring systemfor the grading of direct laryngoscopy. Anaesthesia 1998;53:1041–4.

22. Song F, Khan KS, Dinnes J, Sutton AJ. Asymmetric funnel plotsand publication bias in meta-analyses of diagnostic accuracy. Int

J Epidemiol 2002;31:88–95.23. Practice guidelines for management of the difficult airway: an

updated report by the American Society of AnesthesiologistsTask Force on Management of the Difficult Airway. Anesthesi-ology 2003;98:1269–77.

24. Jaeschke R, Guyatt GH, Sackett DL. Users’ guides to the medical

literature. III. How to use an article about a diagnostic test. B.What are the results and will they help me in caring for mypatients? The Evidence-Based Medicine Working Group. JAMA1994;271:703–7.

25. Shapiro DE. Issues in combining independent estimates of thesensitivity and specificity of a diagnostic test. Acad Radiol.1995;2(Suppl 1):S37–47.

26. Littenberg B, Moses LE. Estimating diagnostic accuracy frommultiple conflicting reports: a new meta-analytic method. MedDecis Making 1993;13:313–21.

27. Moses LE, Shapiro D, Littenberg B. Combining independentstudies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. StatMed 1993;12:1293–316.

28. Jones CM, Athanasiou T. Summary receiver operating charac-teristic curve analysis techniques in the evaluation of diagnostic

tests. Ann Thorac Surg 2005;79:16–20.

29. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuringinconsistency in meta-analyses. BMJ 2003;327:557–60.

30. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.

31. Macaskill P, Walter SD, Irwig L. A comparison of methods todetect publication bias in meta-analysis. Stat Med 2001;20:

641–54.32. Bilgin H, Ozyurt G. Screening tests for predicting difficult

intubation: a clinical assessment in Turkish patients. AnaesthIntensive Care 1998;26:382–6.

33. Cohen SM, Laurito CE, Segil LJ. Examination of the hypophar-ynx predicts ease of laryngoscopic visualization and subsequentintubation: a prospective study of 665 patients. J Clin Anesth1992;4:310–4.

34. Voyagis GS, Kyriakis KP, Dimitriou V, Vrettou I. Value of oropharyngeal Mallampati classification in predicting difficultlaryngoscopy among obese patients. Eur J Anaesthesiol 1998;15:330–4.

35. Tse JC, Rimm EB, Hussain A. Predicting difficult endotrachealintubation in surgical patients scheduled for general anesthesia:a prospective blind study. Anesth Analg 1995;81:254–8.

36. Calder I, Calder J, Crockard HA. Difficult direct laryngoscopy in

patients with cervical spine disease. Anaesthesia 1995;50:756–63.37. Randell T, Maattanen M, Kytta J. The best view at laryngoscopy

using the McCoy laryngoscope with and without cricoid pres-sure. Anaesthesia 1998;53:536–9.

38. Gercek A, Lim S, Isler FB, et al. The prediction of difficultintubation with bedside scoring systems. Marmara Med J 2003;16:16–9.

39. Constantikes J. Predicting difficult tracheal intubation using amodified Mallampati sign: a pilot study report. CRNA 1993;4:16–20.

40. Nadal JL, Fernandez BG, Escobar IC, et al. The palm print as asensitive predictor of difficult laryngoscopy in diabetics. ActaAnaesthesiol Scand 1998;42:199–203.

41. Adnet F, Racine SX, Borron SW, et al. A survey of trachealintubation difficulty in the operating room: a prospective obser-vational study. Acta Anaesthesiol Scand 2001;45:327–32.

42. Bouaggad A, Nejmi SE, Bouderka MA, Abbassi O. Prediction of difficult tracheal intubation in thyroid surgery. Anesth Analg2004;99:603–6.

43. Merah NA, Foulkes-Crabbe DJ, Kushimo OT, Ajayi PA. Predic-tion of difficult laryngoscopy in a population of Nigerian ob-stetric patients. West Afr J Med 2004;23:38–41.

44. Brodsky JB, Lemmens HJM, Brock-Utne JG, et al. Morbid obe-sity and tracheal intubation. Anesth Analg 2002;94:732–6.

45. Ita CE, Eshiet AI, Akpan SG. Recognition of the difficult airwayin normal Nigerian adults (a prospective study). West Afr J Med1994;13:102–4.

46. Gupta S, Pareek S, Dulara SC. Comparison of two methods forpredicting difficult intubation in obstetric patients. Middle East

J Anesthesiol 2003;17:275–85.47. Arne J, Descoins P, Fusciardi J, et al. Preoperative assessment for

difficult intubation in general and ENT surgery: predictive

value of a clinical multivariate risk index. Br J Anaesth 1998;80:140–6.48. Iohom G, Ronayne M, Cunningham AJ. Prediction of difficult

tracheal intubation. Eur J Anaesthesiol 2003;20:31–6.49. Khan ZH, Kashfi A, Ebrahimkhani E. A comparison of the

upper lip bite test (a simple new technique) with modifiedMallampati classification in predicting difficulty in endotra-cheal intubation: a prospective blinded study. Anesth Analg2003;96:595–9.

50. Ayuso MA, Sala X, Luis M, Carbo JM. Predicting difficult oro-tracheal intubation in pharyngo-laryngeal disease: preliminaryresults of a composite index. Can J Anaesth 2003;50:81–5.

51. Schmitt H, Buchfelder M, Radespiel-Troger M, Fahlbusch R.Difficult intubation in acromegalic patients: incidence and pre-dictability. Anesthesiology 2000;93:110–4.

52. Savva D. Prediction of difficult tracheal intubation. Br J Anaesth

1994;73:149–53.




53. Lewis M, Keramati S, Benumof JL, Berry CC. What is the bestway to determine oropharyngeal classification and mandibularspace length to predict difficult laryngoscopy? Anesthesiology1994;81:69–75.

54. Frerk CM. Predicting difficult intubation. Anaesthesia 1991;46:1005–8.

55. Koh LK, Kong CE, Ip-Yam PC. The modified Cormack-Lehane

score for the grading of direct laryngoscopy: evaluation in theAsian population. Anaesth Intensive Care 2002;30:48–51.

56. Kaul TK, Deepika PG Singh A, Bansal M. Prediction of difficultintubation: analysis in 500 adult patients. J Anaesthesiol ClinPharmacol 1998;14:323–8.

57. Ayoub C, Baraka A, el Khatib M, et al. A new cut-off point of thyromental distance for prediction of difficult airway. MiddleEast J Anesthesiol 2000;15:619–33.

58. Jacobsen J, Jensen E, Waldau T, Poulsen TD. Preoperative eval-uation of intubation conditions in patients scheduled for elec-tive surgery. Acta Anaesthesiol Scand 1996;40:421–4.

59. Yeo SW, Chong JL, Thomas E. Difficult intubation: a prospectivestudy. Singapore Med J 1992;33:362–4.

60. Ezri T, Medalion B, Weisenberg M, et al. Increased body massindex per se is not a predictor of difficult laryngoscopy. Can JAnaesth 2003;50:179–83.

61. Ezri T, Gewurtz G, Sessler DI, et al. Prediction of difficultlaryngoscopy in obese patients by ultrasound quantification of anterior neck soft tissue. Anaesthesia 2003;58:1111–4.

62. Ezri T, Weisenberg M, Khazin V, et al. Difficult laryngoscopy:incidence and predictors in patients undergoing coronary artery

bypass surgery versus general surgery patients. J CardiothoracVasc Anesth 2003;17:321–4.

63. Juvin P, Lavaut E, Dupont H, et al. Difficult tracheal intubationis more common in obese than in lean patients. Anesth Analg2003;97:595–600.

64. Vani V, Kamath SK, Naik LD. The palm print as a sensitivepredictor of difficult laryngoscopy in diabetics: a comparisonwith other airway evaluation indices. J Postgrad Med 2000;46:75–9.

65. Langeron O, Masso E, Huraux C, et al. Prediction of difficultmask ventilation. Anesthesiology 2000; 92:1229–36.

66. Egger M, Davey SG, Altman DG. Systematic reviews in healthcare, 2nd ed. London: British Medical Journal Publishing Group,2001.

67. Shiga T, Wajima Z, Inoue T, Sakamoto A. Predicting difficultintubation in apparently normal patients: a meta-analysis of

bedside screening test performance. Anesthesiology 2005;103:429–37.

68. Cohen AM, Fleming BG, Wace JR. Grading of direct laryn-goscopy: a survey of current practice. Anaesthesia 1994;49:522–5.

69. Hawthorne L, Wilson R, Lyons G, Dresner M. Failed intubationrevisited: 17-yr experience in a teaching maternity unit. Br JAnaesth 1996;76:680–4.

70. Samman N, Mohammadi H, Xia J. Cephalometric norms for theupper airway in a healthy Hong Kong Chinese population.Hong Kong Med J 2003;9:25–30.

71. Murphy M, Hung O, Launcelott G, et al. Predicting the difficultlaryngoscopic intubation: are we on the right track? Can J An-aesth 2005;52:231–5.

72. Lijmer JG, Bossuyt PM, Heisterkamp SH. Exploring sources of heterogeneity in systematic reviews of diagnostic tests. Stat Med2002;21:1525–37.

73. Bossuyt PM, Reitsma JB, Bruns DE, et al. Towards complete andaccurate reporting of studies of diagnostic accuracy: the STARD

initiative. BMJ 2003;326:41–4.74. Karkouti K, Rose DK, Ferris LE, et al. Inter-observer reliability

of ten tests used for predicting difficult tracheal intubation. Can J Anaesth 1996;43:554–9.

75. Tham EJ, Gildersleve CD, Sanders LD, et al. Effects of posture,phonation and observer on Mallampati classification. Br J An-aesth 1992;68:32–8.

76. Walter SD. Properties of the summary receiver operating char-acteristic (SROC) curve for diagnostic test data. Stat Med 2002;21:1237–56.

77. Hasselblad V, Hedges LV. Meta-analysis of screening and diag-nostic tests. Psychol Bull 1995;117:167–78.

78. Irwig L, Macaskill P, Glasziou P, Fahey M. Meta-analytic meth-ods for diagnostic test accuracy. J Clin Epidemiol 1995;48:119–30.


mallam pat i

Documents